Python编写接口调用ollama本地部署的大模型,并返回结果

Mac系统,已经通过ollama在本地部署了几个大模型:

通过ollama run qwen:0.5b在后台启动qwen0.5b的模型,然后编写python接口代码,命名为port.py,在里面指定本机接口地址,模型名,输入问题等:

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Mon Oct 21 15:30:00 2024

@author: xiaobai
"""
import requests
 
def llmchat(question):
    host="http://localhost"
    port="11434"
    model = "qwen:0.5b"

    url = f"{host}:{port}/api/chat"
    
    headers = {"Content-Type": "application/json"}

    data = {
            "model": model, 
            "options": {
                "temperature": 0.  
             },
            "stream": False, 
            "messages": [{
                "role": "system",
                "content":question
            }] 
        }

    response=requests.post(url,json=data,headers=headers,timeout=60)
    answer=response.json().get("message").get('content')
    print(answer)
if __name__=="__main__":
    question = "你是谁?" #在这里输入问题
    print('\n')
    llmchat(question)
    

运行port.py程序,大模型返回结果:

作者:福小白

物联沃分享整理
物联沃-IOTWORD物联网 » Python编写接口调用ollama本地部署的大模型,并返回结果

发表回复